Ruhe, GuentherHemmati, Ashkan2018-10-182018-10-182018-10-17Hemmati, A. (2018). Robust Design for Feature Usage-Driven Product Improvement (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/33214http://hdl.handle.net/1880/108893Software product improvement is a multi-layered area of study in software engineering, consisting of various processes and attributes. Innovations in technology, changes in customer and market requirements, as well as variations in user requirements introduce challenges, making it difficult to improve a product and satisfy all stakeholders. These concerns become crucial as more features are offered and resources become scarce. To survive in today’s competitive market, product creators need to do more than just keep their customers satisfied. They must also keep them excited for additions and improvements in future releases. The goal is to increase the overall product value; improve product quality and deliver new features to customers. To find an equilibrium between cost and value, and efficiently deploy scarce resources, we can analyze product improvement with respect to system utilization. In this thesis, an exploratory analytical approach is proposed which incorporates robust experimental designs to build the Robust Predictive Performance Model (RPPM) for feature improvement. Robust experimental designs were first introduced by Dr. Genichi Taguchi in the field of manufacturing to increase the robustness of products or processes. Robustness reduces the impact of uncontrollable factors or noises acting on the product and brings the product or process under statistical control. RPPM applies Taguchi designs in software engineering and extends it to generate a reliable predictive linear regression model, using feature usage as an attribute for quality success, customer satisfaction, and overall product value. These findings can then be analyzed to have an unbiased and reliable evaluation of the product value. As part of this thesis, a case study was performed, applying RPPM on a product created by Calgary-based start-up Brightsquid Inc, a global provider of health communication services to medical and dental professionals. Results for applying RPPM show that this approach can lead to reliable predictions about product value, creating a framework for enhancing decision-making, resource allocations, and the ability to accurately forecast future returns in a frequently changing market.engUniversity of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission.Software Product ImprovementRobust Experimental DesignsExploratory Research in Software EngineeringFeature-Usage AnalysisProduct Performance Predictive ModelSoftware EngineeringComputer Sciencemaster thesis10.11575/PRISM/33214